1. Field
Certain aspects of the present disclosure generally relate to signal processing in wireless body area networks and, more particularly, to a method and an apparatus for mitigating artifacts by using multiple wireless sensors.
2. Background
Biophysical signals from a human body can be sensed by using various sensors such as: Ag/AgCl electrodes for sensing electro-cardiogram (ECG) signals, pulse oximeters for sensing blood oxygenation (SpO2), micro-electro-mechanical (MEM) devices such as 3D-accelerometers for sensing physical activity, and so on. These signals can be then sent wirelessly to a body worn gateway device with connectivity to Wide Area Networks (WAN). It can be possible for automatic programs and/or healthcare providers to interpret vital signs from these biophysical signals, and to provide feedback to the user or intervene in therapy, as appropriate. In such applications, it is beneficial that the signals are free of sensing and transmission artifacts, as such artifacts can impede accurate interpretation.
There are algorithms in the art that perform artifact detection and removal. All of these algorithms detect anomalies in the biophysical signals and perform certain time-frequency domain operations to mitigate the impact of artifacts. However, the artifacts are removed directly in the signal space and not in a compressed sensing (CS) space. The CS may provide a lower implementation complexity, while accuracy of signal interpretation may be improved. For example, the CS acquisition may not produce aliasing due to high frequency components arising from motion and noise.